NUR 314: Final Exam Study Guide
Phi or Cramer's V
A chi square tells us wether or not there are group differences. In order to look at the strength of those differences, what test would we need to run?
True
A correlation of +1 indicates a "perfect" positive relationship, whereas a correlation of -1 indicates a "perfect" negative relationship.
Stratified random sampling
A method of sampling that involves the division of a population into smaller sub-groups known as strata. The strata are formed based on members' shared attributes or characteristics such as income or educational attainment.
Cluster sampling
A probability sampling technique where researchers divide the population into multiple groups for research. Researchers then select random groups with a simple random or systematic random sampling technique for data collection and data analysis
False
A statistically significant result will always be practically important as well.
Simple random sampling
A type of sample taken so that all subjects in a population have an equal probability of being selected
Systematic sampling
A type of sampling that begins with assigning a number to each subject in the population and then selecting every kth person (where k is the population size divided by the desired sample size)
Convenience sampling
A type of sampling that is based primarily on accessibility of subjects in the population
Volunteer sampling
A type of sampling that is taken in a way that only those who offer themselves as participants in the study are included as a sample
Snowball sampling
A type of sampling that uses word of mouth, nomination, or referral to accrue subjects. The researcher contacts one subject, and then they nominate or refer to an additional subject, and the sample begins to grow.
Allows you to adjust for continuous covariates
ANCOVA is different from one-way ANOVA in that ANCOVA
False
All nonparametric tests assume that the data follows a normal distribution.
Yes
Are the alpha and the level of significance are the same thing?
True
Assume that one found an odds ratio of 3.69 for a study looking at an association between exercise (Yes or No) and health status (Good or Poor). This means that the odds of having good health were 3.69 times higher for those who exercise.
Mann-Whitney test with Bonferroni correction
Assume that the results of a Kruskal-Wallis test reached statistical significance. Which of the following can be used as a follow-up test? - Mann-Whitney test with Bonferroni correction - Wilcoxon signed-rank test - Wilcoxon rank-sum test - Mann-Whitney test and Wilcoxon rank-sum test
True
Correlation coefficients may be used as an effect size.
False
Correlation implies causation.
False
Distributional assumption such as normality can be ignored, as most statistical tests are designed to be robust against violation.
True
Fisher's exact test can be used when the sample size is so small that the assumption of the chi-square test is violated.
If skewness is not close to zero, then your data set is not normally distributed
How can skewness be used to determine normality?
By increasing our sample size
How can we control type I and type II errors?
Look for the column labeled "sig", if the data value is under .05, then it is significant There is one exception to this- that is when a Levene's is ran, and if the data value is over .05 then it is significant
How do you identify SPSS output for significance?
2
How many non-parametric equivalents are there for Pearson's correlation?
68% of observations fall within 1 standard deviation away from the mean to both directions
How many standard deviations away from the mean is a 68% confidence level?
95% of observations fall within 2 standard deviations away from the mean to both directions
How many standard deviations away from the mean is a 95% confidence level?
99.7% of observations fall within 3 standard deviations away from the mean to both directions
How many standard deviations away from the mean is a 99% confidence level?
When we're looking at out output for correlation, we are going to first determine if there is a statistically significant relationship - If we do not have a statistically significant relationship (p value is greater than .05), there is no statistically significant relationship between those variables, and therefore we would fail to reject the null hypothesis. We would stop interpreting at this point. - If we do have a statistically significant relationship (p value is less than .05), therefore, we would reject the null hypothesis and continue interpreting. Once we find that there is a statistically significant relationship, we would look at the direction and strength of that correlation - Direction: correlation coefficient = r, if it is negative or positive then that is the direction - Strength: correlation coefficient = r, review table of data values that determines the strength of the coefficient
How to read and interpret SPSS output and identifying SPSS output for significance, direction, and strength of the correlation?
Bonferroni correction (post hoc or planned contrast)
If an ANOVA tells us that at least two group means differ, what do we need to run in order to see where the differences are?
There is almost no variability in the distribution
If you calculate the standard deviation of the data and obtain a value close to 0, which of the following is true? - Standard deviation is miscalculated - There is lots of variability in the distribution - There is almost no variability in the distribution - The mean will be a useless measure of central tendency
There is almost no variability in the distribution
If you calculate the standard deviation of the data and obtain a value close to 0, which of the following is true? - Standard deviation is miscalculated. - There is lots of variability in the distribution. - There is almost no variability in the distribution. - The mean will be a useless measure of central tendency.
True
Nonparametric tests are typically better than the corresponding parametric tests in detecting statistically significant differences when the distribution is not normal.
False
Nonparametric tests require more assumptions than parametric tests.
True
Nonparametric tests should be considered instead of parametric tests when the sample size is small and will lead to a violation of normality assumption.
False
Obtaining a p-value less than the specified alpha in a chi-square test indicates that the two variables are independent.
Standard deviation and range
Of the following, which of the following measures of dispersion is most affected by an outlier? - Standard deviation - Interquartile range - Range - Standard deviation and range
Type 1 error
Repeating t-tests over and over again, we increase our chance of committing a ____________.
Quota sampling
Sampling that is carried out by dividing the population into mutually exclusive (not overlapping) groups and selecting subjects from each group
False
Statistical significance indicates clinical importance.
False
The Friedman's test is best used when the data are continuous and normally distributed.
2; independent
The Mann-Whitney test is used to compare ______ sets of scores that are _______.
True
The chi-square test of association is used in a correlational sense, looking for an association between the two categorical variables.
State the expected relationship between the variables
The hypotheses of the study should:
We compare the mean for each group
The independent t-test is chosen when
True
The level of measurement should be carefully considered as one chooses the best way to graphically present the data.
mean
The null hypothesis of a paired t-test states there is no difference between the ______ of the two groups.
30
The researcher finds data are not exactly normally distributed. However, if there are at least ________ pairs, a dependent t-test could still be used.
Levene's
To test for homogeneity of variance between the groups in a one-way ANOVA, the ______ test is used.
False
True or False: A bar chart is usually good for displaying continuous data.
True
True or False: A distribution is said to be non-normal when both measures of skewness and kurtosis do not fall between -1 and 1
True
True or False: A distribution is said to be normal when both measures of skewness and kurtosis fall between -1 and 1
True
True or False: A scatterplot is a good chart to display information about how the two continuous variables are related.
True
True or False: A test can be reliable without being valid, but it cannot be valid without being reliable
True
True or False: All types of research, from exploratory to experimental, may be analyzed using the same statistical approach.
False
True or False: An ANOVA can tell us where the differences are
True
True or False: Categorical variables can also be called dichotomous variables
True
True or False: Correlation does not equal causation
False
True or False: Correlation equals causation
True
True or False: Evidence-based practice is clinical decision making, by well-informed, expert clinicians, using the best evidence available in the context of individual patient preferences.
True
True or False: For every parametric test, there is a non-parametric equivalent
False
True or False: If we're going to write a directional alternative hypothesis, it is a two-tailed test.
True
True or False: In the case of determining the most effective treatment for a specific pressure ulcer, meta-analyses provide the strongest evidence of cause and effect.
True
True or False: Nominal and ordinal data are considered qualitative
True
True or False: The level of significance of .05 is the most commonly used in applied research.
True
True or False: The more confident we are that our data will fall between two point estimates, the wider our confidence interval will be
False
True or False: The null hypothesis states that there is an effect (i.e., relationship or mean difference), while the alternative hypothesis states that there is no effect.
True
True or False: The numerical distance between the 75th and 25th percentiles is called "interquartile range."
True; when one goes up, the other goes down As our chances for a type I error go down, our chances for a type II error go up
True or False: Type I and type II errors are inversely related
True
True or False: We have our research question already written before we start our hypothesis testing
True
True or False: When looking at SPSS, the p-value is the Sig. (2-tailed) box
False
True or False: When we have a normal distribution, our mean, median, and mode should not be equal
True
True or false: Interval and ratio levels of measurement are continuous data
True
True or false: When running a Levene's test, we want our p-value to be greater than .05
False
We can assume a causal relationship between the two variables if the correlation coefficient is greater than .90.
True
We can assume that the variables are "independent" of each other if the correlation coefficient between the scores is close to "0."
Statistics that allow a researcher to generalize about a population based on the results from the sample. Your reject or fail to reject a null hypothesis using inferential statistics.
What are inferential statistics? What are considered inferential statistics?
An extreme score compared to the other observations you are studying; it will change the way our data looks AND effect the mean of our data When we have outliers and the mean is effected, we would use the median or the IQR instead because the mean would no longer be accurate
What are outliers and what do they do to our distribution?
Single values computes from sample data- when we have a 99% CI, we are confident that 99% of the time our data will fall between these two point estimates
What are point estimates?
Involves collecting and analyzing non-numerical data (e.g., text, video, or audio) to understand concepts, opinions, or experiences
What are qualitative studies?
Variables whose data values are nonnumeric
What are qualitative variables?
The process of collecting and analyzing numerical data It can be used to find patterns and averages, make predictions, test causal relationships, and generalize results to wider populations
What are quantitative studies?
Variables whose data values are numeric
What are quantitative variables?
Nominal Ordinal Interval Ratio
What are the levels of measurement?
Mean Median Mode
What are the measures of central tendency?
Range Standard deviation Interquartile range
What are the measures of variance or variability?
Spearman's Rho Kendall's Tau
What are the non-parametric equivalents for Pearson's correlation?
If you do not meet normality, or have variables that are measured on an ordinal scale (likert scale, pain scale, military rank, etc), then a non parametric equivalent would need to be ran
What are the non-parametric tests for correlation?
It is bell-shaped and symmetric Area under a normal curve is equal to 1.00 or 100% 68% of observations fall within 1 standard deviation away from the mean to both directions 95% of observations fall within 2 standard deviations away from the mean to both directions 99.7% of observations fall within 3 standard deviations away from the mean to both directions Normality can be checked visually with a histogram
What are the properties of a normal distribution?
1. State the null and alternative hypothesis 2. Choose the level of significance 3. Propose an appropriate statistical test 4. Checking assumptions of the chosen test 5 and 6. Running the test and finding the critical value 7. Comparing results and making conclusions
What are the steps of hypothesis testing?
Independent t-test Dependent t-test (paired sample) One sample t-test
What are the three types of parametric t-tests?
Cohen's d Pearson's r coefficient Ω²
What are the types of effect size?
Independent t-test Dependent t-test (paired sample) One sample t-test
What are the types of t-tests?
That at least two group means differ, but not where the differences are
What can an ANOVA tell us?
F statistic
What do we get for ANOVA?
T statistic
What do we get for t-tests?
Bonferroni correction
What do we have to run to determine where the differences are for a statistically significant ANOVA?
It allows us to compare two completely different distributions; we are standardizing it When we do this, it means that our mean will always be 0 and our standard deviation will always be 1 which allows to make comparisons with different distributions
What does a z-score transformation allow us to do?
Analysis of covariance This is when we have the variables of interest, but we want to control for another variable that we know will impact our variables of interest, but we are not interested in that (Ex. is there a difference between the different treatment types on patient outcomes while controlling for patient anxiety)
What is ANCOVA?
When we're looking at examining the differences between three groups
What is ANOVA?
Ranges from 0-1; We would like it to be about .8, or 80%
What is Cohen's d?
For an independent t-test and a one way ANOVA we need to make sure that we meet the assumption of homogeneity of variances (equal variances are assumed) Our p-value for Levene's needs to be GREATER than .05, because that means we did meet the assumption that equal variances are assumed - If we do not meet this assumption, we continue to interpret our independent t-test, but we read off of a different line
What is Levene's test for homogeneity of variance?
When we have Pearson's correlation we know that we have two continuous variables They have met the assumption of normality (bc this is a parametric test) We're looking at correlation to see if there is a relationship between these two variables If you do not meet normality, or have variables that are measured on an ordinal scale (likert scale, pain scale, military rank, etc), then a non parametric equivalent would need to be ran
What is Pearson's correlation?
Repeated measures ANOVA When we are looking at comparing individuals on different time points (Ex. we want to know if a certain exercise prevention is decreasing the BMI of our participants and we want to check them pre-intervention, 3 months after the intervention, 6 months after the intervention, etc)
What is a RANOVA?
Helps decrease type 1 error because it is able to compare multiple groups (multiple comparisons at one time)
What is a bonferroni correction? What does it tell us?
When we look at a relationship between categorical variables Both variables are categorical or dichotomous (Ex. Do you have the disease? Yes or No. Was the treatment effective? Yes or No)
What is a chi square? When do we use this?
When we have a variable that we want to control for; don't necessarily want to examine, but may still impact the outcome of a study - SPSS can do this for us Ex. We are looking at whether or not an intervention decreases BMI, but we want to control for gender so that it doesn't impact the outcome
What is a covariant?
Also known as categorical variable; a variable that can be counted, but has a finite number of countable categories; variables measured at the nominal and ordinal level of measurementI.e., whole numbers, people in a classroom (you can't have half a person)
What is a discrete variable?
If hypotheses imply direction on top of possible differences, then the test is said to be one-tailed. What we use when we know the direction of our data Example alternative hypothesis: Newly developed approach will be more effective than an existing approach in preventing fall incidence.
What is a one-tailed test?
When we want to run a pre-test and post-test
What is a paired samples or dependent t-test?
An entire group of individuals that a researcher wants to study
What is a population
A subset of a population from which a research draws conclusions that are used to understand the population
What is a sample?
When hypotheses are written with no direction of difference, the test is said to be two-tailed.We don't know the direction of our data Example alternative hypothesis: There is a difference in fall incidence between the newly developed approach and an existing approach.
What is a two-tailed test?
Occurs when the null hypothesis is rejected when it should not be
What is a type I error?
Occurs when the null hypothesis is not rejected when it should be
What is a type II error?
Always .05, this means we are at a 95% confidence level 95% of the time we're confident that we are not committing a type I error
What is alpha?
The measure of the strength or magnitude of an effect, difference or relationship, between variables
What is an effect size?
Relationship between the two variables The comparison of covariance is not possible Both variables should be measured on a continuous level of measurement and be normally distributed It is a standardized measure so it allows us to compare coefficients It ranges between -1 and +1
What is bivariate correlation coefficient?
The degree to which an instrument or tool measures the specific idea of interest
What is construct validity?
The degree to which a measurement tool captures the elements of the concept of interest
What is content validity?
A variable that has an infinite number of possible values (ie., every value on a continuum) or the infinite number of values between two consecutive values; variables measured at the interval and ratio level
What is continuous data?
A statistical tool that is used to determine a relationship between the movement between two variables
What is covariance?
The degree to which measurements from one tool or instrument may be correlated with measurements from other valid and reliable instruments
What is criterion-related validity?
When we describe our distribution. Statistics that summarize the data from a sample, instrument, or scale such as central tendency and variation.
What is descriptive statistics?
The validity of a study based on whether or not its results can be generalized from the sample to the target population
What is external validity?
The degree to which changes on the dependent variable may be attributed to the independent variable. Strongly influenced by the quality of the study and control of confounding variables
What is internal validity?
The difference between the values at the 75th percentile and the 25th percentile of a data set
What is interquartile range?
The ability of a test or scale to provide consistent values when used by different people
What is interrater reliability?
A measure of the tailedness of a distribution
What is kurtosis?
Evidence from a systematic review or meta-analysis of all relevant randomized controlled trials, or evidence-based clinical practice guidelines based on systematic reviews of randomized control trials
What is level I evidence?
Evidence obtained from at least one well designed RCT
What is level II evidence?
Evidence obtained from well-designed controlled trials without randomization
What is level III evidence?
Evidence from well-designed case-control and cohort studies
What is level IV evidence?
Evidence from systematic reviews of descriptive and qualitative studies
What is level V evidence?
Evidence from a single descriptive or qualitative study
What is level VI evidence?
Evidence from the opinion of authorities and/or reports of expert committees
What is level VII evidence?
Non-parametric tests are alternatives to parametric tests when the assumptions are violated They do not meet the assumptions of parametric tests
What is non-parametric?
Spearman's rho Kendall's tau Both can be used when parametric assumptions of pearson correlation coefficients are not met (such as non-normally distributed variables or ordinal variables)
What is nonparametric correlation coefficient?
Our data is normally distribution; if we violate this we have to run the non-parametric test
What is one assumption for every parametric test?
Parametric tests have a certain set of assumptions, such a normality When these are violated, the results cannot be trusted
What is parametric?
When there may be other variables that are influencing the main relationship under investigation, it is difficult to examine the true relationship between the two variables Partial correlation coefficient allows to examine the true relationship after controlling for the influence of a third/unwanted variable E.g. relationship between treatment and patient outcome after controlling for patient anxiety
What is partial correlation coefficient?
It can give us a lot of information We can use it for power, we can use a post hoc correction such as bonferroni
What is post-hoc analysis? What information does it provide?
This is conducted after the study is completed and tells you what level of power the study was conducted at with the obtained sample size along with other factors
What is post-hoc power analysis?
A procedure used to calculate the minimum sample size required to be able to detect statistical significance based on effect size, or to calculate the level of power, given a sample size
What is power analysis?
The probability of avoiding a type II error
What is power in layman's terms?
This is completed before the study is conducted and is a guide for estimating the sample size required to reject the null hypothesis when we should
What is priori power analysis?
The best form of sampling, decreases bias, but not always possible When you select people for a sample and every person has the same chance of being selected
What is random sampling?
The difference between the largest and smallest values in a data set
What is range?
The incidence in the people exposed to the risk factor divided by the incidence in the people not exposed to the risk factor
What is relative risk?
The measure of whether or not a test is able to consistently measure a given variable
What is reliability?
A measure of how symmetrical a distribution is
What is skewness?
The average amount that data values will vary from the mean; the square root of the variance
What is standard deviation?
The measure of a test's ability to consistently provide the same measurements across time
What is test-retest reliability?
The data value in the box labeled Pearson Correlation
What is the "r" in the pearson correlation when looking at output from SPSS?
Usually denoted as H1 or Ha; The hypothesis that observations from a sample are influenced by a non random element; the hypothesis the researcher is interested in- there is a statistically significant effect on the variable(s) being studied
What is the alternative hypothesis?
Histograms
What is the appropriate graph for displaying interval data?
Pie charts or bar charts
What is the appropriate graph for displaying nominal data?
Pie charts or bar charts
What is the appropriate graph for displaying ordinal data?
Histograms
What is the appropriate graph for displaying ratio data?
The variable that is affected by the variable that we manipulate Ex. If we are studying the effects of blood pressure medication on blood pressure, the blood pressure would be our DV
What is the dependent variable?
The variable that we as researchers can manipulate Ex. If we are studying the effects of blood pressure medication on blood pressure, the blood pressure medication would be our IV
What is the independent variable?
Data are classified into categories with rankings and are mutually exclusive as in ordinal level of measurement. In addition, specific meanings are applied to the distances between categories (these distances are equal between categories and can be measured). In healthcare, we often use interval level of measurement for clinical tests. Ex. Items that have a measurable distance between them, but no meaningful (non-arbitrary) zero point, such as Fahrenheit and Celsius temperatures.
What is the interval level of measurement?
Always .05, this means we are at a 95% confidence level 95% of the time we're confident that we are not committing a type I error
What is the level of significance?
Average
What is the mean?
The middle number of the data set in numerical order
What is the median?
The number that occurs most often in a data set
What is the mode?
Data are classified into mutually exclusive categories (selection of only one category is allowed) where no ranking or ordering is imposed on categories Ex. a set of items that can be distinguished by name or category; jersey numbers in basketball
What is the nominal level of measurement?
This is a trick question A chi square is a non-parametric test It tells us that there are group differences, but to look at the strength of those differences we would need to run a phi or cramer's v
What is the non-parametric test if assumptions are violated in a chi square?
Donated as H0; The hypothesis that suggests there will be no statistically significant effect on the variable(s) being studied- you either reject or fail to reject the null hypothesis
What is the null hypothesis?
It quantifies the strength of an association between two events Ex. the odds of a person with a disease was exposed to that disease in the past (or exposed to the risk factor) divided by the odds that the control group had to the risk factor
What is the odds ratio?
Data is also classified into mutually exclusive categories, however, ranking or ordering is imposed on categories Ex. Items that can be ordered, such as military rank, or units of government, but whose degree of difference can't be measured.
What is the ordinal level of measurement?
All characteristics of interval level of measurement are present, in addition there is a meaningful zero, and ratio or equal proportion is present. Therefore, it is possible to apply statistical tests with ratio level of measurement that are not possible with the other levels. This is important in health care because income, blood pressure, age, height, and weight are all ratio levels of measurement. Ex. Measurements that have a meaningful zero and can be divided meaningfully, such as the Kelvin temperature scale.
What is the ratio level of measurement?
The dependent t-test
What is the repeated measures ANOVA an extension of?
The extent to which a test measures the variable it is designed to measure
What is validity?
Continuous variable and ratio
What level of measurement is blood pressure?
Ratio
What level of measurement is breaths per minute?
Nominal
What level of measurement is this? A set of items that can be distinguished by name or category.
Ordinal
What level of measurement is this? Items that can be ordered, such as military rank, or units of government, but whose degree of difference can't be measured.
Interval
What level of measurement is this? Items that have a measurable distance between them, but no meaningful (non-arbitrary) zero point, such as Fahrenheit and Celsius temperatures.
Ratio
What level of measurement is this? Measurements that have a meaningful zero and can be divided meaningfully, such as the Kelvin temperature scale.
Data are measured at the interval level
What levels of measurement are the variables in a dependent (paired sample) t-test?
Dependent variable should be continuous and normally distributed
What levels of measurement are the variables in a one sample t-test?
Data should be measured at least at an interval level
What levels of measurement are the variables in an independent t-test?
Friedman's test
What non-parametric test would be used for RANOVA?
Wilcoxon Signed Rank Test
What non-parametric test would be used for dependent t-test?
Mann-Whitney
What non-parametric test would be used for independent t-test?
Kruskal-Wallis
What non-parametric test would be used for one-way ANOVA?
40%
What percentage of the variability among systolic blood pressure (SBP) values is associated with variability among age if the correlation coefficient between SBP and age is 0.60? - 40% - 36% - 60% - 25%
#4
What step of hypothesis testing is this? Checking assumptions of the chosen test
#2
What step of hypothesis testing is this? Choose the level of significance
#7
What step of hypothesis testing is this? Comparing results and making conclusions
#3
What step of hypothesis testing is this? Propose an appropriate statistical test
#5 and #6
What step of hypothesis testing is this? Running the test and finding the critical value
#1
What step of hypothesis testing is this? State the null and alternative hypothesis
No relationship
What strength would a correlation coefficient of 0 - 0.1 be?
Weak relationship
What strength would a correlation coefficient of 0.1 - 0.3 be?
Moderate relationship
What strength would a correlation coefficient of 0.3 - 0.5 be?
Strong relationship
What strength would a correlation coefficient of 0.5 - 0.8 be?
Very strong relationship
What strength would a correlation coefficient of >0.8 be?
Independent t-test (2 groups, if we add a variable to that then we have 3 groups which would mean we would need to run an ANOVA instead)
What t-test is ANOVA an extension of?
Paired samples or dependent t-test
What t-test is RANOVA an extension of?
One-way ANOVA
What test is ANCOVA an extension of?
Independent t-test and one way ANOVA, so far We have to make sure we meet the assumption of variability
What tests do we use for variability?
Two-tailed
What type of hypothesis testing should the investigator use when the direction of an effect is unknown?
Nominal and ordinal variables
What variables are also called categorical variables?
When we have paired data Pre test and post test Ex. I ask a patient if they are aware of the side effects of the medication they are about to take, I give them education, and then later I test them again to see if their education on the medication improved
When do we use a dependent (paired sample) t-test?
When you have a known population mean and you want to compare your sample with that known population mean Ex. IQ scores of first semester nursing students compared to MWSU students' IQs as a whole
When do we use a one sample t-test?
When we are looking at two independent groups Ex. When we are looking at two genders and we want to see of these two genders, is there a difference in salary?
When do we use an independent t-test?
When we have two groups that we want to compareWe know that depending on the groups we have, it will determine what type of t-test we need to use
When do we use t-tests?
Median
Which measure of central tendency is reported when there are significant outliers?
IQR
Which measure of variability is reported when there are significant outliers?
A priori power analysis
Which of the following analyses will help in determining an adequate sample size? - Post hoc power analysis - Interaction analysis - A priori power analysis - Interaction analysis and a priori power analysis
The grouping variable has three or more categories. The variable measuring the characteristic of interest is normally distributed.
Which of the following are assumptions of a one-way ANOVA? (select all that apply)- There is heterogeneity of variance- The grouping variable has three or more categories.- There is homogeneity of variance.- The variable measuring the characteristic of interest is normally distributed.- The measures of the characteristic of interest constitute a dependent sample.
Partial correlation coefficient
Which of the following correlation coefficients allows us to look at the true relationship between the two variables after controlling for the influence of a third/unwanted variable? - Bivariate correlation coefficient - Partial correlation coefficient - Nonparametric correlation coefficient - Bivariate correlation coefficient and nonparametric correlation coefficient
0.05
Which of the following correlation coefficients is the "lowest"? -0.14 0.20 0.05 -0.41
All of these are correct
Which of the following is an important factor to consider when you are deciding what format to display the data?- Level of measurement- Amount of the data- Characteristics of the targeted audience- All of these are correct.
Pie chart
Which of the following is not a good chart to display continuous data? - Histogram - Box and whisker plot - Pie chart - None of these are correct.
The variables should be normally distributed.
Which of the following is not an assumption for the chi-square test? - All observations are independent. - The variables should be normally distributed. - Expected count or cases in each cell should be greater than 1. - No more than 20% of cells should be less than 5.
Is there a gender difference on the average systolic blood pressure?
Which of the following is not an example of research questions that one can answer with a chi-square test? - Is there a difference on preference between four hospitals according to gender? - Is whether or not a person exercises related to whether a person smokes? - Is satisfaction level related to types of therapy patients received? - Is there a gender difference on the average systolic blood pressure?
Wilcoxon signed-rank test
Which of the following is the counterpart of the dependent samples t-test to compare the means of two dependent populations? - Mann-Whitney test - Wilcoxon signed-rank test - Friedman's test - Kruskal-Wallis test
The confidence interval will get wider
Which of the following is true when choosing a higher confidence level in constructing a confidence interval? - The confidence interval will get wider. - It won't affect the confidence interval. - The confidence interval will get narrower. - None of these are correct.
Mean
Which of the following measures of central tendency is affected by an outlier? - Median - Mean - Mode - All of these are correct.
They are more powerful and because we can use continuous data
Why do we prefer parametric tests?
When we're looking at sample size and to make sure our test is powerful enough that we're able to get statistically significant results and not commit a type II error. Power is related to a type II error
Why is power analysis important?